Synopsis

Keywords: Data Acquisition, Lung, 19F fluorine

We studied the feasibility of multi breath wash-in/out lung acquisition with ¹⁹F gas and a sub 0.5 second acquisition spiral sequence. We compared the sub 0.5 seconds acquired data ability to measure wash-in/out time constants to a regular breath-hold ¹⁹F lung imaging and report correlation as well as differences in mean. Our results yielded high correlations (r > 0.88) for both wash-in and wash-out time constant calculations. The results indicate our proposed method yield similar calculated time wash-in/out time constants without the need for breath-holds and show promise for the use of free breathing ¹⁹F gas lung MRI.

Introduction

Analysis of multi-breath wash-in wash-out (MBWW) MRI has been shown to provide a marker of lung disease more sensitive than the current gold standard percent predicted forced expiratory volume (ppFEV1)¹. However, MBWW commonly requires patients to hold their breaths during the scans, which makes the study of children (<5 years) and patient with high disease progression difficult. ¹⁹F has features that make it suitable for MBWW study over other commonly used contrast agents like ¹²⁹Xe and ³He such as it being used without hyperpolarization and it being storable at large volumes premixed in a normoxing mixture of 21% O2, 79 % ¹⁹F. In this abstract we study the feasibility of sub 0.5 seconds (suitable for free breathing) scans during MBWW.

Methods

MRI images were acquired using a multinuclear-capable Prisma 3.0T scanner (Siemens Medical Systems, Erlangen, Germany) and a custom 8-channel transmit/receive ¹⁹F-tuned chest coil (ScanMed, Inc., Omaha NE). 3 healthy and 8 Cystic Fibrosis (CF) individuals participated in this study. Percent predicted forced expiratory volume (ppFEV1) was measured for each participant prior to MR scanning on the same day.¹H MRI scans were acquired for each participant using 3D Fast Low Angle (FL3D) based ultra-short echo time (UTE) while breath-holding for later masking ¹⁹F images (TE/TR 0.05/2.42 ms, FA 5^o, resolution 2.14X2.14x2.5 mm, number of slices 103, scan time 17 seconds).
For ¹⁹F acquisition, the participants underwent 10 wash-in breaths, they were instructed to perform deep inhalations every other breath and breath-hold for the duration of the scans. During the breath-hold, a Volumetric interpolated breath-hold examination (VIBE) sequence (TE/TR 1.61/13 ms, FA 74^o, resolution 6.25X6.25x15 mm, number of slices 18, scan time 18 seconds) followed by a 4 arm spiral sequence images (TE/TR 0.48/11 ms, FA 74^o, resolution 6.25X6.25x15 mm, number of slices 18, scan time 0.45 seconds) were acquired. Wash-out imaging was performed in a similar way while participant’s inspiration gas was returned to room air until no 19F signal was visible.
The VIBE ¹⁹F Maximum intensity image (MIP) was registered to the ¹H UTE and a mask was drawn over the thoracic cavity. The spiral images were denoised² and registered to the ¹⁹F images (translation only). The previously drawn lung mask was used to analyze both ¹⁹F VIBE and ¹⁹F spiral images. The mean signal inside the whole lung was computed for both ¹⁹F method and fitted to a biexponential curve to calculate the wash-in time constant (Tau1) and wash-out time constant (Tau1). MBWW of the lung mean was compared between the two methods for signal intensity, and time constants using Bland-Altman plots.

Results

Good agreement was found between the ¹⁹F VIBE and ¹⁹F Spiral MBWW analysis. Table 1 shows the participants CF state. Figure 1 shows an example VIBE and Spiral coronal screenshots with all participant aggregated mean signals at the bottom of each method. Signal intensity had a high correlation coefficient (r) value of 0.99 (difference in mean of 0.007, p=0.01, Figure 2), the MBWW time constants had r values of 0.83 (Tau1, difference in mean of -1.16, p=0.70, Figure 3) and 0.87 (Tau2, difference in mean of -1.54, p=0.11, Figure 4).

Discussion

The spiral scans are seen to be linearly proportional in signal to the VIBE scans based on Figure 2. Z-score normalization was used due to the denoising process changing the absolute magnitudes. The r values were good for the MBWW time constants Tau1 (r = 0.83) and tau2 (r = 0.87) which show promise of performing MBWW with only the spiral scans. The Bland-Altman difference in mean values were small; -1.16 (Tau1, p = 0.70) and -1.54 (Tau2, p = 0.11), strengthening the feasibility for very short breath-holds or free breathing ¹⁹F dynamic lung MRI.
The low number of subjects is a limitation of this study; we are currently recruiting additional subjects and expect more data points to yield significant time constant p-values. Further analysis is planned for comparison of MBWW metrics on a voxel-by-voxel basis, which will give us both temporal as well as spatial information. Looking forward, we anticipate that further comparisons between full inhalation and tidal volume breathing are also needed.

Conclusion

Good agreement was shown between ¹⁹F MBWW comparison between the breath-hold requiring VIBE sequence (scan time 18 seconds) to the short spiral sequence (scan time 0.45 seconds). MBWW parameters shower good correlation (signal r=0.99, wash-in Tau1 r=0.83, and wash-out r=0.87), suggesting the utility of this approach for the quantitative assessment of lung ventilation.

Acknowledgements

This work was supported by the Cystic Fibrosis Foundation (PROMISE-MCC18K0; GORALS19Y5) and the National Institute of Health (K23-HL138257-01A1; P30 DK065988)